debian/copyright: use spaces rather than tabs to start continuation lines.

[youtubedl] / youtube_dl / extractor / vrv.py

diff --git a/youtube_dl/extractor/vrv.py b/youtube_dl/extractor/vrv.py
index 9959627c0ad3690e02594bc10b9633c023604f7e..6e51469b094363879749bf2926714662e26a1858 100644 (file)
--- a/youtube_dl/extractor/vrv.py
+++ b/youtube_dl/extractor/vrv.py
@@ -11,10 +11,12 @@ import time
 
 from .common import InfoExtractor
 from ..compat import (
 
 from .common import InfoExtractor
 from ..compat import (

+    compat_HTTPError,

     compat_urllib_parse_urlencode,
     compat_urllib_parse_urlencode,

-    compat_urlparse,
+    compat_urllib_parse,

 )
 from ..utils import (
 )
 from ..utils import (

+    ExtractorError,

     float_or_none,
     int_or_none,
 )
     float_or_none,
     int_or_none,
 )


@@ -24,55 +26,78 @@ class VRVBaseIE(InfoExtractor):
     _API_DOMAIN = None
     _API_PARAMS = {}
     _CMS_SIGNING = {}
     _API_DOMAIN = None
     _API_PARAMS = {}
     _CMS_SIGNING = {}

+    _TOKEN = None
+    _TOKEN_SECRET = ''

 
     def _call_api(self, path, video_id, note, data=None):
 
     def _call_api(self, path, video_id, note, data=None):

+        # https://tools.ietf.org/html/rfc5849#section-3

         base_url = self._API_DOMAIN + '/core/' + path
         base_url = self._API_DOMAIN + '/core/' + path

-        encoded_query = compat_urllib_parse_urlencode({
-            'oauth_consumer_key': self._API_PARAMS['oAuthKey'],
-            'oauth_nonce': ''.join([random.choice(string.ascii_letters) for _ in range(32)]),
-            'oauth_signature_method': 'HMAC-SHA1',
-            'oauth_timestamp': int(time.time()),
-            'oauth_version': '1.0',
-        })
+        query = [
+            ('oauth_consumer_key', self._API_PARAMS['oAuthKey']),
+            ('oauth_nonce', ''.join([random.choice(string.ascii_letters) for _ in range(32)])),
+            ('oauth_signature_method', 'HMAC-SHA1'),
+            ('oauth_timestamp', int(time.time())),
+        ]
+        if self._TOKEN:
+            query.append(('oauth_token', self._TOKEN))
+        encoded_query = compat_urllib_parse_urlencode(query)

         headers = self.geo_verification_headers()
         if data:
             data = json.dumps(data).encode()
             headers['Content-Type'] = 'application/json'
         headers = self.geo_verification_headers()
         if data:
             data = json.dumps(data).encode()
             headers['Content-Type'] = 'application/json'

-        method = 'POST' if data else 'GET'
-        base_string = '&'.join([method, compat_urlparse.quote(base_url, ''), compat_urlparse.quote(encoded_query, '')])
+        base_string = '&'.join([
+            'POST' if data else 'GET',
+            compat_urllib_parse.quote(base_url, ''),
+            compat_urllib_parse.quote(encoded_query, '')])

         oauth_signature = base64.b64encode(hmac.new(
         oauth_signature = base64.b64encode(hmac.new(

-            (self._API_PARAMS['oAuthSecret'] + '&').encode('ascii'),
+            (self._API_PARAMS['oAuthSecret'] + '&' + self._TOKEN_SECRET).encode('ascii'),

             base_string.encode(), hashlib.sha1).digest()).decode()
             base_string.encode(), hashlib.sha1).digest()).decode()

-        encoded_query += '&oauth_signature=' + compat_urlparse.quote(oauth_signature, '')
-        return self._download_json(
-            '?'.join([base_url, encoded_query]), video_id,
-            note='Downloading %s JSON metadata' % note, headers=headers, data=data)
+        encoded_query += '&oauth_signature=' + compat_urllib_parse.quote(oauth_signature, '')
+        try:
+            return self._download_json(
+                '?'.join([base_url, encoded_query]), video_id,
+                note='Downloading %s JSON metadata' % note, headers=headers, data=data)
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
+                raise ExtractorError(json.loads(e.cause.read().decode())['message'], expected=True)
+            raise

 
     def _call_cms(self, path, video_id, note):
         if not self._CMS_SIGNING:
 
     def _call_cms(self, path, video_id, note):
         if not self._CMS_SIGNING:

-            self._CMS_SIGNING = self._call_api('index', video_id, 'CMS Signing')['cms_signing']
+            index = self._call_api('index', video_id, 'CMS Signing')
+            self._CMS_SIGNING = index.get('cms_signing') or {}
+            if not self._CMS_SIGNING:
+                for signing_policy in index.get('signing_policies', []):
+                    signing_path = signing_policy.get('path')
+                    if signing_path and signing_path.startswith('/cms/'):
+                        name, value = signing_policy.get('name'), signing_policy.get('value')
+                        if name and value:
+                            self._CMS_SIGNING[name] = value

         return self._download_json(
             self._API_DOMAIN + path, video_id, query=self._CMS_SIGNING,
             note='Downloading %s JSON metadata' % note, headers=self.geo_verification_headers())
 
         return self._download_json(
             self._API_DOMAIN + path, video_id, query=self._CMS_SIGNING,
             note='Downloading %s JSON metadata' % note, headers=self.geo_verification_headers())
 

-    def _set_api_params(self, webpage, video_id):
-        if not self._API_PARAMS:
-            self._API_PARAMS = self._parse_json(self._search_regex(
-                r'window\.__APP_CONFIG__\s*=\s*({.+?})</script>',
-                webpage, 'api config'), video_id)['cxApiParams']
-            self._API_DOMAIN = self._API_PARAMS.get('apiDomain', 'https://api.vrv.co')
-

     def _get_cms_resource(self, resource_key, video_id):
         return self._call_api(
             'cms_resource', video_id, 'resource path', data={
                 'resource_key': resource_key,
             })['__links__']['cms_resource']['href']
 
     def _get_cms_resource(self, resource_key, video_id):
         return self._call_api(
             'cms_resource', video_id, 'resource path', data={
                 'resource_key': resource_key,
             })['__links__']['cms_resource']['href']
 

+    def _real_initialize(self):
+        webpage = self._download_webpage(
+            'https://vrv.co/', None, headers=self.geo_verification_headers())
+        self._API_PARAMS = self._parse_json(self._search_regex(
+            [
+                r'window\.__APP_CONFIG__\s*=\s*({.+?})(?:</script>|;)',
+                r'window\.__APP_CONFIG__\s*=\s*({.+})'
+            ], webpage, 'app config'), None)['cxApiParams']
+        self._API_DOMAIN = self._API_PARAMS.get('apiDomain', 'https://api.vrv.co')
+

 
 class VRVIE(VRVBaseIE):
     IE_NAME = 'vrv'
     _VALID_URL = r'https?://(?:www\.)?vrv\.co/watch/(?P<id>[A-Z0-9]+)'
 
 class VRVIE(VRVBaseIE):
     IE_NAME = 'vrv'
     _VALID_URL = r'https?://(?:www\.)?vrv\.co/watch/(?P<id>[A-Z0-9]+)'

-    _TEST = {
+    _TESTS = [{

         'url': 'https://vrv.co/watch/GR9PNZ396/Hidden-America-with-Jonah-Ray:BOSTON-WHERE-THE-PAST-IS-THE-PRESENT',
         'info_dict': {
             'id': 'GR9PNZ396',
         'url': 'https://vrv.co/watch/GR9PNZ396/Hidden-America-with-Jonah-Ray:BOSTON-WHERE-THE-PAST-IS-THE-PRESENT',
         'info_dict': {
             'id': 'GR9PNZ396',


@@ -85,67 +110,111 @@ class VRVIE(VRVBaseIE):
             # m3u8 download
             'skip_download': True,
         },
             # m3u8 download
             'skip_download': True,
         },

-    }
+    }, {
+        # movie listing
+        'url': 'https://vrv.co/watch/G6NQXZ1J6/Lily-CAT',
+        'info_dict': {
+            'id': 'G6NQXZ1J6',
+            'title': 'Lily C.A.T',
+            'description': 'md5:988b031e7809a6aeb60968be4af7db07',
+        },
+        'playlist_count': 2,
+    }]
+    _NETRC_MACHINE = 'vrv'
+
+    def _real_initialize(self):
+        super(VRVIE, self)._real_initialize()
+
+        email, password = self._get_login_info()
+        if email is None:
+            return
+
+        token_credentials = self._call_api(
+            'authenticate/by:credentials', None, 'Token Credentials', data={
+                'email': email,
+                'password': password,
+            })
+        self._TOKEN = token_credentials['oauth_token']
+        self._TOKEN_SECRET = token_credentials['oauth_token_secret']
+
+    def _extract_vrv_formats(self, url, video_id, stream_format, audio_lang, hardsub_lang):
+        if not url or stream_format not in ('hls', 'dash', 'adaptive_hls'):
+            return []
+        stream_id_list = []
+        if audio_lang:
+            stream_id_list.append('audio-%s' % audio_lang)
+        if hardsub_lang:
+            stream_id_list.append('hardsub-%s' % hardsub_lang)
+        format_id = stream_format
+        if stream_id_list:
+            format_id += '-' + '-'.join(stream_id_list)
+        if 'hls' in stream_format:
+            adaptive_formats = self._extract_m3u8_formats(
+                url, video_id, 'mp4', m3u8_id=format_id,
+                note='Downloading %s information' % format_id,
+                fatal=False)
+        elif stream_format == 'dash':
+            adaptive_formats = self._extract_mpd_formats(
+                url, video_id, mpd_id=format_id,
+                note='Downloading %s information' % format_id,
+                fatal=False)
+        if audio_lang:
+            for f in adaptive_formats:
+                if f.get('acodec') != 'none':
+                    f['language'] = audio_lang
+        return adaptive_formats

 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

-        webpage = self._download_webpage(
-            url, video_id,
-            headers=self.geo_verification_headers())
-        media_resource = self._parse_json(self._search_regex(
-            r'window\.__INITIAL_STATE__\s*=\s*({.+?})</script>',
-            webpage, 'inital state'), video_id).get('watch', {}).get('mediaResource') or {}
-
-        video_data = media_resource.get('json')
-        if not video_data:
-            self._set_api_params(webpage, video_id)
-            episode_path = self._get_cms_resource(
-                'cms:/episodes/' + video_id, video_id)
-            video_data = self._call_cms(episode_path, video_id, 'video')
+
+        object_data = self._call_cms(self._get_cms_resource(
+            'cms:/objects/' + video_id, video_id), video_id, 'object')['items'][0]
+        resource_path = object_data['__links__']['resource']['href']
+        video_data = self._call_cms(resource_path, video_id, 'video')

         title = video_data['title']
         title = video_data['title']

+        description = video_data.get('description')
+
+        if video_data.get('__class__') == 'movie_listing':
+            items = self._call_cms(
+                video_data['__links__']['movie_listing/movies']['href'],
+                video_id, 'movie listing').get('items') or []
+            if len(items) != 1:
+                entries = []
+                for item in items:
+                    item_id = item.get('id')
+                    if not item_id:
+                        continue
+                    entries.append(self.url_result(
+                        'https://vrv.co/watch/' + item_id,
+                        self.ie_key(), item_id, item.get('title')))
+                return self.playlist_result(entries, video_id, title, description)
+            video_data = items[0]

 
 

-        streams_json = media_resource.get('streams', {}).get('json', {})
-        if not streams_json:
-            self._set_api_params(webpage, video_id)
-            streams_path = video_data['__links__']['streams']['href']
-            streams_json = self._call_cms(streams_path, video_id, 'streams')
+        streams_path = video_data['__links__'].get('streams', {}).get('href')
+        if not streams_path:
+            self.raise_login_required()
+        streams_json = self._call_cms(streams_path, video_id, 'streams')

 
         audio_locale = streams_json.get('audio_locale')
         formats = []
         for stream_type, streams in streams_json.get('streams', {}).items():
             if stream_type in ('adaptive_hls', 'adaptive_dash'):
                 for stream in streams.values():
 
         audio_locale = streams_json.get('audio_locale')
         formats = []
         for stream_type, streams in streams_json.get('streams', {}).items():
             if stream_type in ('adaptive_hls', 'adaptive_dash'):
                 for stream in streams.values():

-                    stream_url = stream.get('url')
-                    if not stream_url:
-                        continue
-                    stream_id = stream.get('hardsub_locale') or audio_locale
-                    format_id = '%s-%s' % (stream_type.split('_')[1], stream_id)
-                    if stream_type == 'adaptive_hls':
-                        adaptive_formats = self._extract_m3u8_formats(
-                            stream_url, video_id, 'mp4', m3u8_id=format_id,
-                            note='Downloading %s m3u8 information' % stream_id,
-                            fatal=False)
-                    else:
-                        adaptive_formats = self._extract_mpd_formats(
-                            stream_url, video_id, mpd_id=format_id,
-                            note='Downloading %s MPD information' % stream_id,
-                            fatal=False)
-                    if audio_locale:
-                        for f in adaptive_formats:
-                            if f.get('acodec') != 'none':
-                                f['language'] = audio_locale
-                    formats.extend(adaptive_formats)
+                    formats.extend(self._extract_vrv_formats(
+                        stream.get('url'), video_id, stream_type.split('_')[1],
+                        audio_locale, stream.get('hardsub_locale')))

         self._sort_formats(formats)
 
         subtitles = {}
         self._sort_formats(formats)
 
         subtitles = {}

-        for subtitle in streams_json.get('subtitles', {}).values():
-            subtitle_url = subtitle.get('url')
-            if not subtitle_url:
-                continue
-            subtitles.setdefault(subtitle.get('locale', 'en-US'), []).append({
-                'url': subtitle_url,
-                'ext': subtitle.get('format', 'ass'),
-            })
+        for k in ('captions', 'subtitles'):
+            for subtitle in streams_json.get(k, {}).values():
+                subtitle_url = subtitle.get('url')
+                if not subtitle_url:
+                    continue
+                subtitles.setdefault(subtitle.get('locale', 'en-US'), []).append({
+                    'url': subtitle_url,
+                    'ext': subtitle.get('format', 'ass'),
+                })

 
         thumbnails = []
         for thumbnail in video_data.get('images', {}).get('thumbnails', []):
 
         thumbnails = []
         for thumbnail in video_data.get('images', {}).get('thumbnails', []):


@@ -164,7 +233,7 @@ class VRVIE(VRVBaseIE):
             'formats': formats,
             'subtitles': subtitles,
             'thumbnails': thumbnails,
             'formats': formats,
             'subtitles': subtitles,
             'thumbnails': thumbnails,

-            'description': video_data.get('description'),
+            'description': description,

             'duration': float_or_none(video_data.get('duration_ms'), 1000),
             'uploader_id': video_data.get('channel_id'),
             'series': video_data.get('series_title'),
             'duration': float_or_none(video_data.get('duration_ms'), 1000),
             'uploader_id': video_data.get('channel_id'),
             'series': video_data.get('series_title'),


@@ -190,11 +259,7 @@ class VRVSeriesIE(VRVBaseIE):
 
     def _real_extract(self, url):
         series_id = self._match_id(url)
 
     def _real_extract(self, url):
         series_id = self._match_id(url)

-        webpage = self._download_webpage(
-            url, series_id,
-            headers=self.geo_verification_headers())

 
 

-        self._set_api_params(webpage, series_id)

         seasons_path = self._get_cms_resource(
             'cms:/seasons?series_id=' + series_id, series_id)
         seasons_data = self._call_cms(seasons_path, series_id, 'seasons')
         seasons_path = self._get_cms_resource(
             'cms:/seasons?series_id=' + series_id, series_id)
         seasons_data = self._call_cms(seasons_path, series_id, 'seasons')